The following visualizations are part of my analysis examining the state of American democracy in the 21st century. The data comes from the Vdem Institute’s Varieties of Democracy report from 2021.

Question of Interest:

How has the rise in social media its role in spreading misinformation affected democracy in the United States?

The US is now a ‘Flawed Democracy’

demo_types = c(v2x_polyarchy ="elections are free and fair",v2x_libdem ="individual rights protected", v2x_partipdem ="citizens actively participate\nin all political processes",v2x_delibdem ="political decisions made\n for the common good", v2x_egaldem="rights and freedoms protected\nequally across social groups")

ggplot(vdem %>% filter(country_name == "United States of America") %>% 
            dplyr::select(country_name, year, v2x_polyarchy, v2x_libdem,v2x_partipdem,v2x_delibdem,v2x_egaldem) %>%
               gather(key = "index_type", value = "index_value",3:7), aes(x = year, y = index_value), group = 1) + 
  geom_line(size = 1) + 
  xlim(1980,2021) +
  facet_wrap(~index_type,labeller = labeller(index_type = demo_types)) + 
  labs(title = "The United States is Now a 'Flawed Democracy'", 
       subtitle = "with the most backsliding between 2015 and 2020",
       y = "Democratic index value (0 - 1)", 
       x = "Year") + 
  annotate("rect", fill = "red", alpha = 0.5, 
           xmin = 2015, xmax = 2020,
           ymin = -Inf, ymax = Inf) + 
  theme_bw() + 
  theme(strip.text = element_text(size = 10, face = "italic"), 
        title = element_text(size = 15, face = "bold"), 
        axis.title.y = element_text(size = 10, face = "bold"), 
        axis.title.x = element_text(size = 10, face = "bold"),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(face = "plain", hjust = 0.5),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"))
## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## Warning: Removed 191 rows containing missing values (`geom_line()`).

The most significant Decline in US has been of Deliberative Democracy

ggplot(vdem %>% filter(country_name == "United States of America") %>% 
            dplyr::select(country_name, year, v2x_delibdem), aes(x = year, y = v2x_delibdem), group = 1) + 
  geom_line(size = 1) + 
  xlim(2014,2021) + 
  labs(title = "The Deliberative Democracy Rating\nin USA Dropped by 28 % in 5 Years", 
       subtitle = "",
       y = "Deliberative Democracy Rating", 
       x = "Year") + 
  annotate("rect", fill = "red", alpha = 0.5, 
           xmin = 2015, xmax = 2020,
           ymin = -Inf, ymax = Inf) + 
  theme_bw() + 
  theme(strip.text = element_text(size = 10, face = "italic"), 
        title = element_text(size = 20, face = "bold"), 
        axis.title.y = element_text(size = 10, face = "bold"), 
        axis.title.x = element_text(size = 10, face = "bold"),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(face = "plain", hjust = 0.5),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1")) + 
  annotate("label", x = 2015, y = 0.81, label = "0.85", fontface = "bold", size = 7, alpha =.5) + 
  annotate("label", x = 2020, y = 0.54, label = "0.57", fontface = "bold", size = 7, alpha =.5)
## Warning: Removed 225 rows containing missing values (`geom_line()`).

Deliberative Democracy vs. Political Misinformation:

“The deliberative principle of democracy focuses on the process by which decisions are reached in a polity. A deliberative process is one in which public reasoning focused on the common good motivates political decisions—as contrasted with emotional appeals, solidary attachments, parochial interests, or coercion.”

#arranging countries by worst change in democratic rating 
dem_slide = vdem %>% filter(year == 2000 | year == 2010 | year == 2020) %>% 
            dplyr::select(country_name, year, v2x_polyarchy) %>%
              spread(key = year, value = v2x_polyarchy) %>% 
                mutate(d_diff = `2020` - `2000`) %>% 
                  arrange(d_diff)
top_40 = dem_slide[1:40,]
worst_change40 = top_40$country_name 

#USA dataframe 
usa = vdem %>% filter(country_name == "United States of America" & (year == 2000 | year == 2010 | year == 2020)) %>% 
  dplyr::select(country_name, year, v2smpardom, v2x_delibdem)

#Creating USA label 
usa_text = data.frame(v2smpardom = 1.8, v2x_delibdem = .70, lab = "USA", year = 2000)

#Plotting graphs
vdem %>% filter(country_name %in% worst_change40 & ( year == 2000 | year == 2010 | year == 2020)) %>% 
            dplyr::select(country_name, year, v2smpardom, v2x_delibdem) %>% 
              ggplot(aes(x = v2smpardom, y = v2x_delibdem), color = "grey") + geom_point(alpha =.5) + 
              geom_point(data = usa, aes(x = v2smpardom, y = v2x_delibdem), color = "red", size = 5, shape = 15 ) +
              scale_x_reverse(labels = c('Extremely Often', 'Often', 'About half the time', 'Rarely', 'Never, or almost never'), limits = c(3,-3)) +
              geom_text(data = usa_text, label = "USA", color = "red", fontface = "bold") +
              theme_bw() + 
              theme(
                    plot.title = element_text(face = "bold", size = 18, hjust = 0.5), 
                    plot.subtitle = element_text(face = "plain", size = 15, hjust = 0.5),
                    plot.caption = element_text(face = "plain", size = 8),
                    plot.background = element_rect(fill = "cornsilk1"),
                    legend.background = element_rect(fill = "cornsilk1")) + 
              facet_grid(year~.) + 
              labs(title = "USA is Ranked 32nd Among the 40 Countries\nwith the Most Decline in Deliberative Democracy", 
                   subtitle = "The Countries with the Most Decline Are Bolivia, Venezuela, & Nicaragua", 
                   x = "Goverment use of Social Media to Spread Misinformation", 
                   y = "Deliberative Democracy Rating", 
                   caption = "Source: Varieties of Democracy Codebook v12")

Animated version

dem_slide = vdem %>% filter(year == 2000 | year == 2010 | year == 2020) %>%
            dplyr::select(country_name, year, v2x_polyarchy) %>%
              spread(key = year, value = v2x_polyarchy) %>%
                mutate(d_diff = `2020` - `2000`) %>%
                  arrange(d_diff)
top_40 = dem_slide[1:40,]
worst_change40 = top_40$country_name


#USA dataframe
usa = vdem %>% filter(country_name == "United States of America" & (year %in% 2000:2021)) %>%
  dplyr::select(country_name, year, v2smpardom, v2x_delibdem)

#Creating USA label
usa_text = data.frame(v2smpardom = 1.8, v2x_delibdem = .70, lab = "USA", year = 2000:2021)
usa_text
#Plotting graphs

countries2 = vdem %>% filter(country_name %in% worst_change40 & ( year %in% 2000:2021)) %>%
            dplyr::select(country_name, year, v2smpardom, v2x_delibdem)

p = ggplot(countries2, aes(label = country_name, x = v2smpardom, y = v2x_delibdem, frame = year)) + 
              geom_point(alpha =.8, color = "grey") +
              geom_point(data = usa, aes(x = v2smpardom, y = v2x_delibdem, frame=year), color = "red", size = 5, shape = 15 ) +
              scale_x_reverse(labels = c('Extremely Often', 'Often', 'About half the time', 'Rarely', 'Never, or almost never'), limits = c(3,-3)) +
              geom_text(data = usa, aes(x = v2smpardom, y = v2x_delibdem, frame=year), label = "USA", color = "red", fontface = "bold", nudge_x = .1, nudge_y = .1) +
              theme_bw() +
              theme(
                    plot.title = element_text(face = "bold", size = 10, hjust = 0.5),
                    plot.subtitle = element_text(face = "plain", size = 15, hjust = 0.5),
                    plot.caption = element_text(face = "plain", size = 8),
                    plot.background = element_rect(fill = "cornsilk1"),
                    legend.background = element_rect(fill = "cornsilk1")) +
              labs(title = "Democracy Weakens Globally as Politicians Use Social Media\nto Spread False and Misleading Information",
                   subtitle = "USA is among the 40 Countries with Worst Deliberative Democratic Decline from 2000 to 2021",
                   x = "How often major political parties and candidates use social media to spread false information",
                   y = "Deliberative Democracy Rating")
## Warning in geom_point(data = usa, aes(x = v2smpardom, y = v2x_delibdem, :
## Ignoring unknown aesthetics: frame
## Warning in geom_text(data = usa, aes(x = v2smpardom, y = v2x_delibdem, frame =
## year), : Ignoring unknown aesthetics: frame
ggplotly(p)

Example of Social Media’s Rise in Influence: Facebook

html=read_html('https://www.statista.com/statistics/264810/number-of-monthly-active-facebook-users-worldwide/')
table=html_table(html)
facebook=as.data.frame(table)
faceq4=facebook[grep("Q4",facebook$Characteristic),]

faceq4$Characteristic=sub("Q4 '","20", faceq4$Characteristic)
faceq4$Number.of.users.in.millions=sub(",","",faceq4$Number.of.users.in.millions)

faceq4=faceq4 %>% rename("Year"="Characteristic","MAU_Millions"='Number.of.users.in.millions')
faceq4=faceq4 %>% mutate(Year=as.numeric(Year),MAU_Millions=as.numeric(MAU_Millions))
faceq4=faceq4 %>% mutate(MAU_Billions=MAU_Millions/1000)

ggplot(faceq4, aes(x = Year)) + 
 geom_line(aes(y = MAU_Billions), color = "red", size = 2)+
  xlim(2008,2021) +
  scale_y_continuous(breaks=c(0,.5,1,1.5,2,2.5,3))+
  scale_x_continuous(breaks=c(2009,2011,2013,2015,2017,2019,2021))+
  ylab("Monthly Active Users (Billions)")+
  labs(title = "Facebook Popularity Growth")+
  theme_bw() + 
  theme(title = element_text(size = 10, face = "bold"),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"))
## Scale for x is already present.
## Adding another scale for x, which will replace the existing scale.

Social Media is Becoming a Political Propaganda Machine in USA

usa = vdem %>% filter(country_name == "United States of America") 

ggplot(usa, aes(x = year)) + 
  geom_line(aes(y = v2smpardom_osp), color = "forestgreen", size = 2) + 
  geom_line(aes(y = v2smgovdom_osp), color = "purple", size =2) + 
  
  xlim(1999,2021) + 
  scale_y_reverse(labels = c('Extremely Often', 'Often', 'About half the time', 'Rarely', 'Never, or almost never'), limits = c(4,0)) + 
  labs(title = "Social Media is Becoming\na Political Propaganda Machine in USA",
       y = "Frequency of False Info dissemination", x = "Year") + 
  
  annotate("text", x = 2005, y = 2.4, label = "False information spread by\npolitical parties and candidates\non social media", color = "forestgreen", size = 4, fontface = "bold") + 
  annotate("text", x = 2015, y = 2, label = "False information\nspread by government\nofficials on social media  ", color = "purple", size = 4, fontface = "bold") + 
  theme_bw() + 
  theme(title = element_text(size = 15, face = "bold"),
        plot.title = element_text(hjust = 0.5),
        axis.title.y = element_text(size = 10),
        axis.title.x = element_text(size = 10),
        plot.subtitle = element_text(hjust = 0.5),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"))
## Warning: Removed 211 rows containing missing values (`geom_line()`).
## Removed 211 rows containing missing values (`geom_line()`).

Our Political Discourse is Influenced by Foreign Misinformation Campaigns

ggplot(usa, aes(x = year), group = 1) + 
  geom_line(aes(y = v2smfordom_osp), color = "blue", size = 2) + 
  geom_line(aes(y = v2smforads_osp), color = "red", size = 2) + 
  scale_y_reverse(labels = c('Extremely Often', 'Often', 'About half the time', 'Rarely', 'Never, or almost never'), limits = c(4,0)) +
  xlim(2000,2021) + 
  labs(title = "Our Political Discourse is Influenced\nby Foreign Disinformation Campaigns", 
       y = "Frequency of dissemination", x = "Year") + 
  annotate("text", x = 2005, y = 1.5, label = "False information posted on\nsocial media by foreign agents",size =4, color = "blue", fontface = "bold") + 
  annotate("text", x = 2008, y = 3.2, label = "Online advertising bought by foreign agents\n to spread misleading viewpoints",size =4, color = "red", fontface = "bold") + 
  theme_bw() + 
  theme(title = element_text(size = 16, face = "bold"),
        axis.title.y = element_text(size = 10),
        axis.title.x = element_text(size = 10),
        plot.title = element_text(hjust = 0.5),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"))
## Warning: Removed 211 rows containing missing values (`geom_line()`).
## Removed 211 rows containing missing values (`geom_line()`).

US Politics is Becoming More Dangerous

usa2 = vdem %>% filter(country_name == "United States of America") %>% 
  dplyr::select(country_name, year, v2caviol, v2cacamps) %>%
               gather(key = "index_type", value = "index_value", 3:4)

ggplot(usa2, aes( x = year, y = index_value)) + 
  geom_line(aes(color = index_type), size = 2) + 
  xlim(1999,2021) +
  scale_color_manual(values = c("coral1", "cornflowerblue")) + 
  scale_y_continuous(limits=c(-1,3),breaks = c(-1,0,1,2,3), labels=c("Never","Rare","Sometimes","Frequently","Often"))+
  labs(title = "US Politics is Becoming More Dangerous", y="Aggregated Expert Response")+
  annotate("text", x = 2010, y = 1, label = "Political Polarization",size =4, color = "coral1", fontface = "bold") + 
  annotate("text", x = 2013, y = 0, label = "Political Violence",size =4, color = "cornflowerblue", fontface = "bold") +
  theme_bw() + 
  theme(strip.text = element_text(size = 10),
        title = element_text(size = 15, face = "bold"), 
        plot.title = element_text(hjust = 0.5), 
         axis.title.y = element_text(size = 10),
        axis.title.x = element_text(size = 10),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"),
        strip.text.x = element_text(face = "bold"), 
        legend.position = "none")
## Warning: Removed 420 rows containing missing values (`geom_line()`).

Hate Speech & Harassement of Journalists

metrics_of_demo2 = c(v2meharjrn_osp ="Journalist Harassment",v2smpolhate_osp ="Hate Speech")

usa3 = vdem %>% filter(country_name == "United States of America") %>% 
  dplyr::select(country_name, year, v2meharjrn_osp, v2smpolhate_osp) %>%
               gather(key = "index_type", value = "index_value", 3:4)

ggplot(usa3, aes( x = year, y = index_value), group = 1)+
  geom_line(aes(color = index_type), size = 2)+
  xlim(2000,2021)+
  scale_y_reverse(limits=c(4,0), labels=c("Certain","Often","Sometimes","Rare","Never"))+
  scale_color_manual(values = c("coral1", "cornflowerblue")) +
  ggtitle("As Political Social Media Campaigns Flourish, USA is Increasingly Hostile")+
  labs(y="Aggregated Expert Response")+
  annotate("text", x = 2010, y = 2.1, label = "Hate Speech",size =4, color = "cornflowerblue", fontface = "bold") + 
  annotate("text", x = 2017, y = 3.7, label = "Harrassment of Journalists",size =4, color = "coral1", fontface = "bold") +
  theme_bw() + 
  theme(title = element_text(size = 8, face = "bold"), 
        plot.title = element_text(hjust = 0.5),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"),
        strip.text.x = element_text(face = "bold"), 
        legend.position = "none")
## Warning: Removed 422 rows containing missing values (`geom_line()`).

Rise in Misinformation Means More Real World Violence

vdem %>% filter(country_name==c("United States of America")) %>% filter(year>2000) %>% 
  ggplot(aes(x=year,y=-v2smorgviol))+
  geom_point(color = "grey")+
  geom_smooth(se=FALSE, color = "red", size = 2)+
  scale_y_continuous(limits =c(0,2),breaks=c(0,1,2),labels=c("Never","Sometimes","Often"))+
  ggtitle("This Polarization & Hostility on Social Media\nResults in Real World Violence")+
  labs(y="Frequency of Offline Violence Organized on Social Media", x = "Year")+
  
  theme_bw() + 
  theme(title = element_text(size = 15, face = "bold"), 
        plot.title = element_text(hjust = 0.5),
        axis.title.y = element_text(size = 10),
        axis.title.x = element_text(size = 10),
        plot.background = element_rect(fill = "cornsilk1"),
        legend.background = element_rect(fill = "cornsilk1"))
## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'